The Persistent Measurement Biases
Abstract: Despite recent advancements in the field of accounting-based valuation, the challenges in linking forecasts of residual income levels to past financial statements; and simultaneously considering the effects of conservatism in the accounting, remain. This paper describes the exploration of an equity valuation model that adheres to a linear information model as conceptualised by Ohlson (1995), and Feltham and Ohlson (1995) in the forecasts of earnings, which is parsimonious and individual for each firm. I find that such a model, that uses the bias estimates (PMBs) of equity as defined by Runsten (1998) for the Swedish market while not permitting short selling; can achieve positive abnormal returns (AR) for the sample period between 2003 to 2015. The observed AR levels were independent of the market cap, and the P/E or B/M - ratio when controlled for their quintile levels. Furthermore, correct assessments of "under or over"-valuation states were in a 12 months forward-looking period in excess of 50 percent, independent from the benchmark performance, the quintile levels of market cap, and the P/E or B/M - ratio. The findings warrant further research towards the nature and cause of observed AR levels, as well as towards whether the model merely tracks the expected cost of capital or captures short-term mispricings of the Swedish markets.
AT THIS PAGE YOU CAN DOWNLOAD THE WHOLE ESSAY. (follow the link to the next page)